diff options
Diffstat (limited to 'fs/cifs')
-rw-r--r-- | fs/cifs/cifsfs.c | 2 | ||||
-rw-r--r-- | fs/cifs/cifsglob.h | 29 | ||||
-rw-r--r-- | fs/cifs/cifspdu.h | 4 | ||||
-rw-r--r-- | fs/cifs/cifsproto.h | 29 | ||||
-rw-r--r-- | fs/cifs/cifssmb.c | 364 | ||||
-rw-r--r-- | fs/cifs/connect.c | 515 | ||||
-rw-r--r-- | fs/cifs/file.c | 295 | ||||
-rw-r--r-- | fs/cifs/transport.c | 8 |
8 files changed, 880 insertions, 366 deletions
diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 96a48baad8f..f219dccbe15 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -74,7 +74,7 @@ module_param(cifs_min_small, int, 0); MODULE_PARM_DESC(cifs_min_small, "Small network buffers in pool. Default: 30 " "Range: 2 to 256"); unsigned int cifs_max_pending = CIFS_MAX_REQ; -module_param(cifs_max_pending, int, 0); +module_param(cifs_max_pending, int, 0444); MODULE_PARM_DESC(cifs_max_pending, "Simultaneous requests to server. " "Default: 50 Range: 2 to 256"); unsigned short echo_retries = 5; diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h index 55ebf39fb3f..d153d0b89d3 100644 --- a/fs/cifs/cifsglob.h +++ b/fs/cifs/cifsglob.h @@ -291,7 +291,13 @@ struct TCP_Server_Info { bool sec_kerberosu2u; /* supports U2U Kerberos */ bool sec_kerberos; /* supports plain Kerberos */ bool sec_mskerberos; /* supports legacy MS Kerberos */ + bool large_buf; /* is current buffer large? */ struct delayed_work echo; /* echo ping workqueue job */ + struct kvec *iov; /* reusable kvec array for receives */ + unsigned int nr_iov; /* number of kvecs in array */ + char *smallbuf; /* pointer to current "small" buffer */ + char *bigbuf; /* pointer to current "big" buffer */ + unsigned int total_read; /* total amount of data read in this pass */ #ifdef CONFIG_CIFS_FSCACHE struct fscache_cookie *fscache; /* client index cache cookie */ #endif @@ -650,8 +656,24 @@ static inline void cifs_stats_bytes_read(struct cifs_tcon *tcon, struct mid_q_entry; /* - * This is the prototype for the mid callback function. When creating one, - * take special care to avoid deadlocks. Things to bear in mind: + * This is the prototype for the mid receive function. This function is for + * receiving the rest of the SMB frame, starting with the WordCount (which is + * just after the MID in struct smb_hdr). Note: + * + * - This will be called by cifsd, with no locks held. + * - The mid will still be on the pending_mid_q. + * - mid->resp_buf will point to the current buffer. + * + * Returns zero on a successful receive, or an error. The receive state in + * the TCP_Server_Info will also be updated. + */ +typedef int (mid_receive_t)(struct TCP_Server_Info *server, + struct mid_q_entry *mid); + +/* + * This is the prototype for the mid callback function. This is called once the + * mid has been received off of the socket. When creating one, take special + * care to avoid deadlocks. Things to bear in mind: * * - it will be called by cifsd, with no locks held * - the mid will be removed from any lists @@ -669,9 +691,10 @@ struct mid_q_entry { unsigned long when_sent; /* time when smb send finished */ unsigned long when_received; /* when demux complete (taken off wire) */ #endif + mid_receive_t *receive; /* call receive callback */ mid_callback_t *callback; /* call completion callback */ void *callback_data; /* general purpose pointer for callback */ - struct smb_hdr *resp_buf; /* response buffer */ + struct smb_hdr *resp_buf; /* pointer to received SMB header */ int midState; /* wish this were enum but can not pass to wait_event */ __u8 command; /* smb command code */ bool largeBuf:1; /* if valid response, is pointer to large buf */ diff --git a/fs/cifs/cifspdu.h b/fs/cifs/cifspdu.h index de3aa285de0..3c6ef34fe2b 100644 --- a/fs/cifs/cifspdu.h +++ b/fs/cifs/cifspdu.h @@ -1089,9 +1089,7 @@ typedef struct smb_com_read_rsp { __le16 DataLengthHigh; __u64 Reserved2; __u16 ByteCount; - __u8 Pad; /* BB check for whether padded to DWORD - boundary and optimum performance here */ - char Data[1]; + /* read response data immediately follows */ } __attribute__((packed)) READ_RSP; typedef struct locking_andx_range { diff --git a/fs/cifs/cifsproto.h b/fs/cifs/cifsproto.h index a1fa9cec05d..c25d0636cc4 100644 --- a/fs/cifs/cifsproto.h +++ b/fs/cifs/cifsproto.h @@ -69,8 +69,9 @@ extern struct mid_q_entry *AllocMidQEntry(const struct smb_hdr *smb_buffer, struct TCP_Server_Info *server); extern void DeleteMidQEntry(struct mid_q_entry *midEntry); extern int cifs_call_async(struct TCP_Server_Info *server, struct kvec *iov, - unsigned int nvec, mid_callback_t *callback, - void *cbdata, bool ignore_pend); + unsigned int nvec, mid_receive_t *receive, + mid_callback_t *callback, void *cbdata, + bool ignore_pend); extern int SendReceive(const unsigned int /* xid */ , struct cifs_ses *, struct smb_hdr * /* input */ , struct smb_hdr * /* out */ , @@ -153,6 +154,12 @@ extern struct cifs_ntsd *get_cifs_acl(struct cifs_sb_info *, struct inode *, extern int set_cifs_acl(struct cifs_ntsd *, __u32, struct inode *, const char *, int); +extern void dequeue_mid(struct mid_q_entry *mid, bool malformed); +extern int cifs_read_from_socket(struct TCP_Server_Info *server, char *buf, + unsigned int to_read); +extern int cifs_readv_from_socket(struct TCP_Server_Info *server, + struct kvec *iov_orig, unsigned int nr_segs, + unsigned int to_read); extern void cifs_setup_cifs_sb(struct smb_vol *pvolume_info, struct cifs_sb_info *cifs_sb); extern int cifs_match_super(struct super_block *, void *); @@ -442,6 +449,24 @@ extern int E_md4hash(const unsigned char *passwd, unsigned char *p16); extern int SMBencrypt(unsigned char *passwd, const unsigned char *c8, unsigned char *p24); +/* asynchronous read support */ +struct cifs_readdata { + struct cifsFileInfo *cfile; + struct address_space *mapping; + __u64 offset; + unsigned int bytes; + pid_t pid; + int result; + struct list_head pages; + struct work_struct work; + unsigned int nr_iov; + struct kvec iov[1]; +}; + +struct cifs_readdata *cifs_readdata_alloc(unsigned int nr_pages); +void cifs_readdata_free(struct cifs_readdata *rdata); +int cifs_async_readv(struct cifs_readdata *rdata); + /* asynchronous write support */ struct cifs_writedata { struct kref refcount; diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index c824c106b2b..aaad4ce6e6c 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -33,6 +33,8 @@ #include <linux/slab.h> #include <linux/posix_acl_xattr.h> #include <linux/pagemap.h> +#include <linux/swap.h> +#include <linux/task_io_accounting_ops.h> #include <asm/uaccess.h> #include "cifspdu.h" #include "cifsglob.h" @@ -40,6 +42,7 @@ #include "cifsproto.h" #include "cifs_unicode.h" #include "cifs_debug.h" +#include "fscache.h" #ifdef CONFIG_CIFS_POSIX static struct { @@ -83,6 +86,9 @@ static struct { #endif /* CONFIG_CIFS_WEAK_PW_HASH */ #endif /* CIFS_POSIX */ +/* Forward declarations */ +static void cifs_readv_complete(struct work_struct *work); + /* Mark as invalid, all open files on tree connections since they were closed when session to server was lost */ static void mark_open_files_invalid(struct cifs_tcon *pTcon) @@ -737,7 +743,8 @@ CIFSSMBEcho(struct TCP_Server_Info *server) iov.iov_base = smb; iov.iov_len = be32_to_cpu(smb->hdr.smb_buf_length) + 4; - rc = cifs_call_async(server, &iov, 1, cifs_echo_callback, server, true); + rc = cifs_call_async(server, &iov, 1, NULL, cifs_echo_callback, + server, true); if (rc) cFYI(1, "Echo request failed: %d", rc); @@ -1374,6 +1381,359 @@ openRetry: return rc; } +struct cifs_readdata * +cifs_readdata_alloc(unsigned int nr_pages) +{ + struct cifs_readdata *rdata; + + /* readdata + 1 kvec for each page */ + rdata = kzalloc(sizeof(*rdata) + + sizeof(struct kvec) * nr_pages, GFP_KERNEL); + if (rdata != NULL) { + INIT_WORK(&rdata->work, cifs_readv_complete); + INIT_LIST_HEAD(&rdata->pages); + } + return rdata; +} + +void +cifs_readdata_free(struct cifs_readdata *rdata) +{ + cifsFileInfo_put(rdata->cfile); + kfree(rdata); +} + +/* + * Discard any remaining data in the current SMB. To do this, we borrow the + * current bigbuf. + */ +static int +cifs_readv_discard(struct TCP_Server_Info *server, struct mid_q_entry *mid) +{ + READ_RSP *rsp = (READ_RSP *)server->smallbuf; + unsigned int rfclen = be32_to_cpu(rsp->hdr.smb_buf_length); + int remaining = rfclen + 4 - server->total_read; + struct cifs_readdata *rdata = mid->callback_data; + + while (remaining > 0) { + int length; + + length = cifs_read_from_socket(server, server->bigbuf, + min_t(unsigned int, remaining, + CIFSMaxBufSize + MAX_CIFS_HDR_SIZE)); + if (length < 0) + return length; + server->total_read += length; + remaining -= length; + } + + dequeue_mid(mid, rdata->result); + return 0; +} + +static int +cifs_readv_receive(struct TCP_Server_Info *server, struct mid_q_entry *mid) +{ + int length, len; + unsigned int data_offset, remaining, data_len; + struct cifs_readdata *rdata = mid->callback_data; + READ_RSP *rsp = (READ_RSP *)server->smallbuf; + unsigned int rfclen = be32_to_cpu(rsp->hdr.smb_buf_length) + 4; + u64 eof; + pgoff_t eof_index; + struct page *page, *tpage; + + cFYI(1, "%s: mid=%u offset=%llu bytes=%u", __func__, + mid->mid, rdata->offset, rdata->bytes); + + /* + * read the rest of READ_RSP header (sans Data array), or whatever we + * can if there's not enough data. At this point, we've read down to + * the Mid. + */ + len = min_t(unsigned int, rfclen, sizeof(*rsp)) - + sizeof(struct smb_hdr) + 1; + + rdata->iov[0].iov_base = server->smallbuf + sizeof(struct smb_hdr) - 1; + rdata->iov[0].iov_len = len; + + length = cifs_readv_from_socket(server, rdata->iov, 1, len); + if (length < 0) + return length; + server->total_read += length; + + /* Was the SMB read successful? */ + rdata->result = map_smb_to_linux_error(&rsp->hdr, false); + if (rdata->result != 0) { + cFYI(1, "%s: server returned error %d", __func__, + rdata->result); + return cifs_readv_discard(server, mid); + } + + /* Is there enough to get to the rest of the READ_RSP header? */ + if (server->total_read < sizeof(READ_RSP)) { + cFYI(1, "%s: server returned short header. got=%u expected=%zu", + __func__, server->total_read, sizeof(READ_RSP)); + rdata->result = -EIO; + return cifs_readv_discard(server, mid); + } + + data_offset = le16_to_cpu(rsp->DataOffset) + 4; + if (data_offset < server->total_read) { + /* + * win2k8 sometimes sends an offset of 0 when the read + * is beyond the EOF. Treat it as if the data starts just after + * the header. + */ + cFYI(1, "%s: data offset (%u) inside read response header", + __func__, data_offset); + data_offset = server->total_read; + } else if (data_offset > MAX_CIFS_SMALL_BUFFER_SIZE) { + /* data_offset is beyond the end of smallbuf */ + cFYI(1, "%s: data offset (%u) beyond end of smallbuf", + __func__, data_offset); + rdata->result = -EIO; + return cifs_readv_discard(server, mid); + } + + cFYI(1, "%s: total_read=%u data_offset=%u", __func__, + server->total_read, data_offset); + + len = data_offset - server->total_read; + if (len > 0) { + /* read any junk before data into the rest of smallbuf */ + rdata->iov[0].iov_base = server->smallbuf + server->total_read; + rdata->iov[0].iov_len = len; + length = cifs_readv_from_socket(server, rdata->iov, 1, len); + if (length < 0) + return length; + server->total_read += length; + } + + /* set up first iov for signature check */ + rdata->iov[0].iov_base = server->smallbuf; + rdata->iov[0].iov_len = server->total_read; + cFYI(1, "0: iov_base=%p iov_len=%zu", + rdata->iov[0].iov_base, rdata->iov[0].iov_len); + + /* how much data is in the response? */ + data_len = le16_to_cpu(rsp->DataLengthHigh) << 16; + data_len += le16_to_cpu(rsp->DataLength); + if (data_offset + data_len > rfclen) { + /* data_len is corrupt -- discard frame */ + rdata->result = -EIO; + return cifs_readv_discard(server, mid); + } + + /* marshal up the page array */ + len = 0; + remaining = data_len; + rdata->nr_iov = 1; + + /* determine the eof that the server (probably) has */ + eof = CIFS_I(rdata->mapping->host)->server_eof; + eof_index = eof ? (eof - 1) >> PAGE_CACHE_SHIFT : 0; + cFYI(1, "eof=%llu eof_index=%lu", eof, eof_index); + + list_for_each_entry_safe(page, tpage, &rdata->pages, lru) { + if (remaining >= PAGE_CACHE_SIZE) { + /* enough data to fill the page */ + rdata->iov[rdata->nr_iov].iov_base = kmap(page); + rdata->iov[rdata->nr_iov].iov_len = PAGE_CACHE_SIZE; + cFYI(1, "%u: idx=%lu iov_base=%p iov_len=%zu", + rdata->nr_iov, page->index, + rdata->iov[rdata->nr_iov].iov_base, + rdata->iov[rdata->nr_iov].iov_len); + ++rdata->nr_iov; + len += PAGE_CACHE_SIZE; + remaining -= PAGE_CACHE_SIZE; + } else if (remaining > 0) { + /* enough for partial page, fill and zero the rest */ + rdata->iov[rdata->nr_iov].iov_base = kmap(page); + rdata->iov[rdata->nr_iov].iov_len = remaining; + cFYI(1, "%u: idx=%lu iov_base=%p iov_len=%zu", + rdata->nr_iov, page->index, + rdata->iov[rdata->nr_iov].iov_base, + rdata->iov[rdata->nr_iov].iov_len); + memset(rdata->iov[rdata->nr_iov].iov_base + remaining, + '\0', PAGE_CACHE_SIZE - remaining); + ++rdata->nr_iov; + len += remaining; + remaining = 0; + } else if (page->index > eof_index) { + /* + * The VFS will not try to do readahead past the + * i_size, but it's possible that we have outstanding + * writes with gaps in the middle and the i_size hasn't + * caught up yet. Populate those with zeroed out pages + * to prevent the VFS from repeatedly attempting to + * fill them until the writes are flushed. + */ + zero_user(page, 0, PAGE_CACHE_SIZE); + list_del(&page->lru); + lru_cache_add_file(page); + flush_dcache_page(page); + SetPageUptodate(page); + unlock_page(page); + page_cache_release(page); + } else { + /* no need to hold page hostage */ + list_del(&page->lru); + lru_cache_add_file(page); + unlock_page(page); + page_cache_release(page); + } + } + + /* issue the read if we have any iovecs left to fill */ + if (rdata->nr_iov > 1) { + length = cifs_readv_from_socket(server, &rdata->iov[1], + rdata->nr_iov - 1, len); + if (length < 0) + return length; + server->total_read += length; + } else { + length = 0; + } + + rdata->bytes = length; + + cFYI(1, "total_read=%u rfclen=%u remaining=%u", server->total_read, + rfclen, remaining); + + /* discard anything left over */ + if (server->total_read < rfclen) + return cifs_readv_discard(server, mid); + + dequeue_mid(mid, false); + return length; +} + +static void +cifs_readv_complete(struct work_struct *work) +{ + struct cifs_readdata *rdata = container_of(work, + struct cifs_readdata, work); + struct page *page, *tpage; + + list_for_each_entry_safe(page, tpage, &rdata->pages, lru) { + list_del(&page->lru); + lru_cache_add_file(page); + kunmap(page); + + if (rdata->result == 0) { + flush_dcache_page(page); + SetPageUptodate(page); + } + + unlock_page(page); + + if (rdata->result == 0) + cifs_readpage_to_fscache(rdata->mapping->host, page); + + page_cache_release(page); + } + cifs_readdata_free(rdata); +} + +static void +cifs_readv_callback(struct mid_q_entry *mid) +{ + struct cifs_readdata *rdata = mid->callback_data; + struct cifs_tcon *tcon = tlink_tcon(rdata->cfile->tlink); + struct TCP_Server_Info *server = tcon->ses->server; + + cFYI(1, "%s: mid=%u state=%d result=%d bytes=%u", __func__, + mid->mid, mid->midState, rdata->result, rdata->bytes); + + switch (mid->midState) { + case MID_RESPONSE_RECEIVED: + /* result already set, check signature */ + if (server->sec_mode & + (SECMODE_SIGN_REQUIRED | SECMODE_SIGN_ENABLED)) { + if (cifs_verify_signature(rdata->iov, rdata->nr_iov, + server, mid->sequence_number + 1)) + cERROR(1, "Unexpected SMB signature"); + } + /* FIXME: should this be counted toward the initiating task? */ + task_io_account_read(rdata->bytes); + cifs_stats_bytes_read(tcon, rdata->bytes); + break; + case MID_REQUEST_SUBMITTED: + case MID_RETRY_NEEDED: + rdata->result = -EAGAIN; + break; + default: + rdata->result = -EIO; + } + + queue_work(system_nrt_wq, &rdata->work); + DeleteMidQEntry(mid); + atomic_dec(&server->inFlight); + wake_up(&server->request_q); +} + +/* cifs_async_readv - send an async write, and set up mid to handle result */ +int +cifs_async_readv(struct cifs_readdata *rdata) +{ + int rc; + READ_REQ *smb = NULL; + int wct; + struct cifs_tcon *tcon = tlink_tcon(rdata->cfile->tlink); + + cFYI(1, "%s: offset=%llu bytes=%u", __func__, + rdata->offset, rdata->bytes); + + if (tcon->ses->capabilities & CAP_LARGE_FILES) + wct = 12; + else { + wct = 10; /* old style read */ + if ((rdata->offset >> 32) > 0) { + /* can not handle this big offset for old */ + return -EIO; + } + } + + rc = small_smb_init(SMB_COM_READ_ANDX, wct, tcon, (void **)&smb); + if (rc) + return rc; + + smb->hdr.Pid = cpu_to_le16((__u16)rdata->pid); + smb->hdr.PidHigh = cpu_to_le16((__u16)(rdata->pid >> 16)); + + smb->AndXCommand = 0xFF; /* none */ + smb->Fid = rdata->cfile->netfid; + smb->OffsetLow = cpu_to_le32(rdata->offset & 0xFFFFFFFF); + if (wct == 12) + smb->OffsetHigh = cpu_to_le32(rdata->offset >> 32); + smb->Remaining = 0; + smb->MaxCount = cpu_to_le16(rdata->bytes & 0xFFFF); + smb->MaxCountHigh = cpu_to_le32(rdata->bytes >> 16); + if (wct == 12) + smb->ByteCount = 0; + else { + /* old style read */ + struct smb_com_readx_req *smbr = + (struct smb_com_readx_req *)smb; + smbr->ByteCount = 0; + } + + /* 4 for RFC1001 length + 1 for BCC */ + rdata->iov[0].iov_base = smb; + rdata->iov[0].iov_len = be32_to_cpu(smb->hdr.smb_buf_length) + 4; + + rc = cifs_call_async(tcon->ses->server, rdata->iov, 1, + cifs_readv_receive, cifs_readv_callback, + rdata, false); + + if (rc == 0) + cifs_stats_inc(&tcon->num_reads); + + cifs_small_buf_release(smb); + return rc; +} + int CIFSSMBRead(const int xid, struct cifs_io_parms *io_parms, unsigned int *nbytes, char **buf, int *pbuf_type) @@ -1834,7 +2194,7 @@ cifs_async_writev(struct cifs_writedata *wdata) kref_get(&wdata->refcount); rc = cifs_call_async(tcon->ses->server, iov, wdata->nr_pages + 1, - cifs_writev_callback, wdata, false); + NULL, cifs_writev_callback, wdata, false); if (rc == 0) cifs_stats_inc(&tcon->num_writes); diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index 97a65af2a08..f70d87d6ba6 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -320,27 +320,24 @@ requeue_echo: } static bool -allocate_buffers(char **bigbuf, char **smallbuf, unsigned int size, - bool is_large_buf) +allocate_buffers(struct TCP_Server_Info *server) { - char *bbuf = *bigbuf, *sbuf = *smallbuf; - - if (bbuf == NULL) { - bbuf = (char *)cifs_buf_get(); - if (!bbuf) { + if (!server->bigbuf) { + server->bigbuf = (char *)cifs_buf_get(); + if (!server->bigbuf) { cERROR(1, "No memory for large SMB response"); msleep(3000); /* retry will check if exiting */ return false; } - } else if (is_large_buf) { + } else if (server->large_buf) { /* we are reusing a dirty large buf, clear its start */ - memset(bbuf, 0, size); + memset(server->bigbuf, 0, sizeof(struct smb_hdr)); } - if (sbuf == NULL) { - sbuf = (char *)cifs_small_buf_get(); - if (!sbuf) { + if (!server->smallbuf) { + server->smallbuf = (char *)cifs_small_buf_get(); + if (!server->smallbuf) { cERROR(1, "No memory for SMB response"); msleep(1000); /* retry will check if exiting */ @@ -349,12 +346,9 @@ allocate_buffers(char **bigbuf, char **smallbuf, unsigned int size, /* beginning of smb buffer is cleared in our buf_get */ } else { /* if existing small buf clear beginning */ - memset(sbuf, 0, size); + memset(server->smallbuf, 0, sizeof(struct smb_hdr)); } - *bigbuf = bbuf; - *smallbuf = sbuf; - return true; } @@ -375,14 +369,72 @@ server_unresponsive(struct TCP_Server_Info *server) return false; } -static int -read_from_socket(struct TCP_Server_Info *server, char *buf, - unsigned int to_read) +/* + * kvec_array_init - clone a kvec array, and advance into it + * @new: pointer to memory for cloned array + * @iov: pointer to original array + * @nr_segs: number of members in original array + * @bytes: number of bytes to advance into the cloned array + * + * This function will copy the array provided in iov to a section of memory + * and advance the specified number of bytes into the new array. It returns + * the number of segments in the new array. "new" must be at least as big as + * the original iov array. + */ +static unsigned int +kvec_array_init(struct kvec *new, struct kvec *iov, unsigned int nr_segs, + size_t bytes) +{ + size_t base = 0; + + while (bytes || !iov->iov_len) { + int copy = min(bytes, iov->iov_len); + + bytes -= copy; + base += copy; + if (iov->iov_len == base) { + iov++; + nr_segs--; + base = 0; + } + } + memcpy(new, iov, sizeof(*iov) * nr_segs); + new->iov_base += base; + new->iov_len -= base; + return nr_segs; +} + +static struct kvec * +get_server_iovec(struct TCP_Server_Info *server, unsigned int nr_segs) +{ + struct kvec *new_iov; + + if (server->iov && nr_segs <= server->nr_iov) + return server->iov; + + /* not big enough -- allocate a new one and release the old */ + new_iov = kmalloc(sizeof(*new_iov) * nr_segs, GFP_NOFS); + if (new_iov) { + kfree(server->iov); + server->iov = new_iov; + server->nr_iov = nr_segs; + } + return new_iov; +} + +int +cifs_readv_from_socket(struct TCP_Server_Info *server, struct kvec *iov_orig, + unsigned int nr_segs, unsigned int to_read) { int length = 0; int total_read; + unsigned int segs; struct msghdr smb_msg; - struct kvec iov; + struct kvec *iov; + + iov = get_server_iovec(server, nr_segs); + if (!iov) + return -ENOMEM; smb_msg.msg_control = NULL; smb_msg.msg_controllen = 0; @@ -393,10 +445,11 @@ read_from_socket(struct TCP_Server_Info *server, char *buf, break; } - iov.iov_base = buf + total_read; - iov.iov_len = to_read; - length = kernel_recvmsg(server->ssocket, &smb_msg, &iov, 1, - to_read, 0); + segs = kvec_array_init(iov, iov_orig, nr_segs, total_read); + + length = kernel_recvmsg(server->ssocket, &smb_msg, + iov, segs, to_read, 0); + if (server->tcpStatus == CifsExiting) { total_read = -ESHUTDOWN; break; @@ -426,6 +479,18 @@ read_from_socket(struct TCP_Server_Info *server, char *buf, return total_read; } +int +cifs_read_from_socket(struct TCP_Server_Info *server, char *buf, + unsigned int to_read) +{ + struct kvec iov; + + iov.iov_base = buf; + iov.iov_len = to_read; + + return cifs_readv_from_socket(server, &iov, 1, to_read); +} + static bool is_smb_response(struct TCP_Server_Info *server, unsigned char type) { @@ -471,61 +536,76 @@ is_smb_response(struct TCP_Server_Info *server, unsigned char type) } static struct mid_q_entry * -find_cifs_mid(struct TCP_Server_Info *server, struct smb_hdr *buf, - int *length, bool is_large_buf, bool *is_multi_rsp, char **bigbuf) +find_mid(struct TCP_Server_Info *server, struct smb_hdr *buf) { - struct mid_q_entry *mid = NULL, *tmp_mid, *ret = NULL; + struct mid_q_entry *mid; spin_lock(&GlobalMid_Lock); - list_for_each_entry_safe(mid, tmp_mid, &server->pending_mid_q, qhead) { - if (mid->mid != buf->Mid || - mid->midState != MID_REQUEST_SUBMITTED || - mid->command != buf->Command) - continue; - - if (*length == 0 && check2ndT2(buf) > 0) { - /* We have a multipart transact2 resp */ - *is_multi_rsp = true; - if (mid->resp_buf) { - /* merge response - fix up 1st*/ - *length = coalesce_t2(buf, mid->resp_buf); - if (*length > 0) { - *length = 0; - mid->multiRsp = true; - break; - } - /* All parts received or packet is malformed. */ - mid->multiEnd = true; - goto multi_t2_fnd; - } - if (!is_large_buf) { - /*FIXME: switch to already allocated largebuf?*/ - cERROR(1, "1st trans2 resp needs bigbuf"); - } else { - /* Have first buffer */ - mid->resp_buf = buf; - mid->largeBuf = true; - *bigbuf = NULL; - } - break; + list_for_each_entry(mid, &server->pending_mid_q, qhead) { + if (mid->mid == buf->Mid && + mid->midState == MID_REQUEST_SUBMITTED && + mid->command == buf->Command) { + spin_unlock(&GlobalMid_Lock); + return mid; } - mid->resp_buf = buf; - mid->largeBuf = is_large_buf; -multi_t2_fnd: - if (*length == 0) - mid->midState = MID_RESPONSE_RECEIVED; - else - mid->midState = MID_RESPONSE_MALFORMED; + } + spin_unlock(&GlobalMid_Lock); + return NULL; +} + +void +dequeue_mid(struct mid_q_entry *mid, bool malformed) +{ #ifdef CONFIG_CIFS_STATS2 - mid->when_received = jiffies; + mid->when_received = jiffies; #endif - list_del_init(&mid->qhead); - ret = mid; - break; - } + spin_lock(&GlobalMid_Lock); + if (!malformed) + mid->midState = MID_RESPONSE_RECEIVED; + else + mid->midState = MID_RESPONSE_MALFORMED; + list_del_init(&mid->qhead); spin_unlock(&GlobalMid_Lock); +} - return ret; +static void +handle_mid(struct mid_q_entry *mid, struct TCP_Server_Info *server, + struct smb_hdr *buf, int malformed) +{ + if (malformed == 0 && check2ndT2(buf) > 0) { + mid->multiRsp = true; + if (mid->resp_buf) { + /* merge response - fix up 1st*/ + malformed = coalesce_t2(buf, mid->resp_buf); + if (malformed > 0) + return; + + /* All parts received or packet is malformed. */ + mid->multiEnd = true; + return dequeue_mid(mid, malformed); + } + if (!server->large_buf) { + /*FIXME: switch to already allocated largebuf?*/ + cERROR(1, "1st trans2 resp needs bigbuf"); + } else { + /* Have first buffer */ + mid->resp_buf = buf; + mid->largeBuf = true; + server->bigbuf = NULL; + } + return; + } + mid->resp_buf = buf; + mid->largeBuf = server->large_buf; + /* Was previous buf put in mpx struct for multi-rsp? */ + if (!mid->multiRsp) { + /* smb buffer will be freed by user thread */ + if (server->large_buf) + server->bigbuf = NULL; + else + server->smallbuf = NULL; + } + dequeue_mid(mid, malformed); } static void clean_demultiplex_info(struct TCP_Server_Info *server) @@ -615,6 +695,7 @@ static void clean_demultiplex_info(struct TCP_Server_Info *server) } kfree(server->hostname); + kfree(server->iov); kfree(server); length = atomic_dec_return(&tcpSesAllocCount); @@ -624,17 +705,70 @@ static void clean_demultiplex_info(struct TCP_Server_Info *server) } static int +standard_receive3(struct TCP_Server_Info *server, struct mid_q_entry *mid) +{ + int length; + char *buf = server->smallbuf; + struct smb_hdr *smb_buffer = (struct smb_hdr *)buf; + unsigned int pdu_length = be32_to_cpu(smb_buffer->smb_buf_length); + + /* make sure this will fit in a large buffer */ + if (pdu_length > CIFSMaxBufSize + MAX_CIFS_HDR_SIZE - 4) { + cERROR(1, "SMB response too long (%u bytes)", + pdu_length); + cifs_reconnect(server); + wake_up(&server->response_q); + return -EAGAIN; + } + + /* switch to large buffer if too big for a small one */ + if (pdu_length > MAX_CIFS_SMALL_BUFFER_SIZE - 4) { + server->large_buf = true; + memcpy(server->bigbuf, server->smallbuf, server->total_read); + buf = server->bigbuf; + smb_buffer = (struct smb_hdr *)buf; + } + + /* now read the rest */ + length = cifs_read_from_socket(server, + buf + sizeof(struct smb_hdr) - 1, + pdu_length - sizeof(struct smb_hdr) + 1 + 4); + if (length < 0) + return length; + server->total_read += length; + + dump_smb(smb_buffer, server->total_read); + + /* + * We know that we received enough to get to the MID as we + * checked the pdu_length earlier. Now check to see + * if the rest of the header is OK. We borrow the length + * var for the rest of the loop to avoid a new stack var. + * + * 48 bytes is enough to display the header and a little bit + * into the payload for debugging purposes. + */ + length = checkSMB(smb_buffer, smb_buffer->Mid, server->total_read); + if (length != 0) + cifs_dump_mem("Bad SMB: ", buf, + min_t(unsigned int, server->total_read, 48)); + + if (mid) + handle_mid(mid, server, smb_buffer, length); + + return length; +} + +static int cifs_demultiplex_thread(void *p) { int length; struct TCP_Server_Info *server = p; - unsigned int pdu_length, total_read; - char *buf = NULL, *bigbuf = NULL, *smallbuf = NULL; + unsigned int pdu_length; + char *buf = NULL; struct smb_hdr *smb_buffer = NULL; struct task_struct *task_to_wake = NULL; struct mid_q_entry *mid_entry; - bool isLargeBuf = false; - bool isMultiRsp = false; current->flags |= PF_MEMALLOC; cFYI(1, "Demultiplex PID: %d", task_pid_nr(current)); @@ -649,20 +783,18 @@ cifs_demultiplex_thread(void *p) if (try_to_freeze()) continue; - if (!allocate_buffers(&bigbuf, &smallbuf, - sizeof(struct smb_hdr), isLargeBuf)) + if (!allocate_buffers(server)) continue; - isLargeBuf = false; - isMultiRsp = false; - smb_buffer = (struct smb_hdr *)smallbuf; - buf = smallbuf; + server->large_buf = false; + smb_buffer = (struct smb_hdr *)server->smallbuf; + buf = server->smallbuf; pdu_length = 4; /* enough to get RFC1001 header */ - length = read_from_socket(server, buf, pdu_length); + length = cifs_read_from_socket(server, buf, pdu_length); if (length < 0) continue; - total_read = length; + server->total_read = length; /* * The right amount was read from socket - 4 bytes, @@ -674,64 +806,42 @@ cifs_demultiplex_thread(void *p) if (!is_smb_response(server, buf[0])) continue; - /* check the length */ - if ((pdu_length > CIFSMaxBufSize + MAX_CIFS_HDR_SIZE - 4) || - (pdu_length < sizeof(struct smb_hdr) - 1 - 4)) { - cERROR(1, "Invalid size SMB length %d pdu_length %d", - 4, pdu_length + 4); + /* make sure we have enough to get to the MID */ + if (pdu_length < sizeof(struct smb_hdr) - 1 - 4) { + cERROR(1, "SMB response too short (%u bytes)", + pdu_length); cifs_reconnect(server); wake_up(&server->response_q); continue; } - /* else length ok */ - if (pdu_length > MAX_CIFS_SMALL_BUFFER_SIZE - 4) { - isLargeBuf = true; - memcpy(bigbuf, smallbuf, 4); - smb_buffer = (struct smb_hdr *)bigbuf; - buf = bigbuf; - } - - length = read_from_socket(server, buf + 4, pdu_length); + /* read down to the MID */ + length = cifs_read_from_socket(server, buf + 4, + sizeof(struct smb_hdr) - 1 - 4); if (length < 0) continue; - total_read += length; + server->total_read += length; - dump_smb(smb_buffer, total_read); + mid_entry = find_mid(server, smb_buffer); - /* - * We know that we received enough to get to the MID as we - * checked the pdu_length earlier. Now check to see - * if the rest of the header is OK. We borrow the length - * var for the rest of the loop to avoid a new stack var. - * - * 48 bytes is enough to display the header and a little bit - * into the payload for debugging purposes. - */ - length = checkSMB(smb_buffer, smb_buffer->Mid, total_read); - if (length != 0) - cifs_dump_mem("Bad SMB: ", buf, - min_t(unsigned int, total_read, 48)); + if (!mid_entry || !mid_entry->receive) + length = standard_receive3(server, mid_entry); + else + length = mid_entry->receive(server, mid_entry); - server->lstrp = jiffies; + if (length < 0) + continue; - mid_entry = find_cifs_mid(server, smb_buffer, &length, - isLargeBuf, &isMultiRsp, &bigbuf); + if (server->large_buf) { + buf = server->bigbuf; + smb_buffer = (struct smb_hdr *)buf; + } + + server->lstrp = jiffies; if (mid_entry != NULL) { - mid_entry->callback(mid_entry); - /* Was previous buf put in mpx struct for multi-rsp? */ - if (!isMultiRsp) { - /* smb buffer will be freed by user thread */ - if (isLargeBuf) - bigbuf = NULL; - else - smallbuf = NULL; - } - } else if (length != 0) { - /* response sanity checks failed */ - continue; - } else if (!is_valid_oplock_break(smb_buffer, server) && - !isMultiRsp) { + if (!mid_entry->multiRsp || mid_entry->multiEnd) + mid_entry->callback(mid_entry); |